Abstract
Lyme disease (LD) is a tick-borne infection caused by the bacterial pathogen Borrelia burgdorferi. Current diagnostic tests mostly use borrelial lysates or select antigens to detect serum antibodies against B. burgdorferi. These immunoassays are not entirely effective, especially for detection of early infection. We have recently characterized an in vivo-induced antigen, BBK07, as a serodiagnostic marker for LD. We now report that in a line blot assay, recombinant BBK07 protein-based detection is 90% sensitive and nearly 100% specific against B. burgdorferi infection in humans. Using an overlapping peptide library of 23 peptides encompassing full-length BBK07, we identified the immunodominant epitopes of BBK07 during human infection. We show that a select combination of amino-terminal peptides significantly enhanced BBK07-based diagnostic accuracy compared to that with the full-length protein. Although in enzyme-linked immunosorbent assay (ELISA) studies BBK07 peptides had overall lower sensitivity than established serodiagnostic peptides, such as the VlsE peptide C6 and OspC peptide pepC10, for the detection of early human LD, a subset of serum samples that failed to recognize either VlsE or OspC peptides were preferentially reactive to BBK07 peptides. These results highlight the fact that BBK07 peptides could be useful to complement the efficacy of VlsE and OspC peptide-based serodiagnostic assays. Finally, using a panel of canine sera, we show that BBK07 peptide is also effective for LD diagnosis in infected dogs. Together, our data show that peptides from the B. burgdorferi surface protein BBK07 are highly specific and sensitive serodiagnostic markers, and we suggest their future use in LD diagnostic assays.
Lyme disease (LD), which is transmitted by the bites of infected Ixodes ticks, can cause debilitating arthritis and neurological symptoms if not promptly treated. LD, also known as Lyme borreliosis, is a globally emerging disease (29) which results from infection with the spirochetes Borrelia burgdorferi in the United States and Canada and B. burgdorferi, Borrelia afzelii, and Borrelia garinii in Eurasia. Antibiotic therapy is usually effective to cure LD, and a human vaccine to prevent the incidence of LD is currently unavailable. Every year more than 20,000 patients are diagnosed in the United States, and an estimated 2 million LD serological tests are performed (1, 5). The diagnosis of LD is complicated by several factors, including frequently unnoticed exposure and dramatic variation in individual immune responses. Unsuspecting humans often present with a variety of symptoms weeks to months after exposure, making laboratory diagnosis an important tool to aid physicians.
Seroreactivity is currently the primary method of laboratory diagnosis of LD (2). If the patient does not present erythema migrans, a pathognomonic skin rash, a two-tier serological approach for LD diagnosis has been recommended by the Centers for Disease Control and Prevention (3). A tested serum sample that is considered positive or equivocal by a sensitive method, most commonly enzyme-linked immunosorbent assay (ELISA), is then subsequently tested by an immunoblot analysis against B. burgdorferi lysates. The latter method is considered to be relatively more specific and is used for the reduction of false positives which results from antigen cross-reactivity in ELISA-based assays (21). Although ELISA using whole-cell lysates is the most frequently used format for the sensitive first-tier testing (2), the outcome is inherently limited by difficulties in standardization. Additionally, the existence of several genospecies and occurrence of antigenic variability among clinical isolates for certain antigens may limit the effectiveness of whole-cell lysates. The antigenic profile of in vitro-grown B. burgdorferi can also significantly vary by growth phase, number of passages, and composition of the complex culture media, among other factors, resulting in substantial batch-to-batch variation in ELISA kits (30, 34).
With millions of serum samples tested for LD each year, standardization and automation of serological testing are major goals of LD research (2). Recombinant and synthetic antigen ELISA kits ease standardization, are amenable to automation, and may improve specificity by concentrating B. burgdorferi-specific epitopes. Among the synthetic antigens previously tested for early LD diagnosis is the OspC-derived peptide pepC10 (25). Highly conserved among Borrelia strains, pepC10 is 10 amino acids in length and appears to be a target for IgM antibodies during the early infection (4). The synthetic peptide C6, isolated from a conserved region of the variable membrane protein VlsE, is a target for host IgG and has been shown to be a sensitive and specific serodiagnostic marker (4, 22). While purified antigens show great promise, no recombinant or synthetic antigen has demonstrated sufficient sensitivity to replace the current two-tier approach (7, 14, 16, 20, 24, 26, 27, 35). Some of the highest sensitivities reported thus far have been obtained by the use of several antigens in combination to enhance diagnostic accuracy (4, 8, 19, 31, 32). However, there remains a need for improvement in sensitivity, especially for detection during the earliest stages of disease (1). The addition of new immunogenic epitopes could allow these tests to eventually supplant the two-tier approach, improving both the efficacy and cost of LD testing.
The B. burgdorferi lipoprotein BBK07 was identified as an immunodominant antigen in a study by Barbour et al. (6). We have recently shown that BBK07 is an in vivo-induced surface antigen which is selectively expressed during mammalian infection and is a promising serodiagnostic marker for LD (9). We demonstrated that an amino-terminal fragment of BBK07 could be used as a component of an effective serodiagnostic marker to detect human LD. In our current studies we further assessed the sensitivity and specificities of BBK07-based diagnosis using serum samples from North American and European patients with diagnosed LD and several other conditions, including syphilis and autoimmune diseases. We also assessed the serodiagnostic abilities of BBK07 using a full-length protein and an overlapping peptide library, identifying the most immunogenic epitopes of BBK07. We demonstrate that serum testing using a combination of peptides was superior to use of the full-length BBK07 protein. Finally, we show that the peptides are able to detect both canine and human LD, even during the early stages of the disease. IgG and IgM ELISAs further show that a cohort of human LD sera failed to recognize VlsE-derived C6 or OspC-derived pepC10 peptide but reacted exclusively with BBK07 peptides, attesting to their potential use in enhancing the diagnostic sensitivity for early LD.
MATERIALS AND METHODS
Bacteria and mice.
Borrelia burgdorferi isolates B31 A3 (13) and 297 (28) were used in the current study. Borrelia burgdorferi clinical isolates B408, B491, B500, B515, BL203, and BL206 were isolated from human patients as described previously (37). Bacteria were grown in BSK-II medium at 34°C as detailed previously (28). C3H/HeN mice were purchased from the National Cancer Institute, and all animal procedures were performed in compliance with the guidelines and with the approval of the Institutional Animal Care and Use Committee.
Purification of recombinant full-length and truncated BBK07.
Full-length BBK07 protein was produced in insect cells using the Bac-to-Bac baculovirus expression system (Invitrogen). The complete BBK07 open reading frame without the native signal peptide was amplified using the following primers, with restriction sites underlined: forward primer with BamHI site, TTT CGG ATC CAT GCA CCA CCA CCA CCA CCA CTG GCA TGT AGA CAA TCC CAT TG; reverse primer with XhoI site, CCC CCT CGA GAA TTA ATT ATT AAA GCA CAA ATG T. The resulting amplicon was inserted into the pFASTBAC-1 vector. A recombinant bacmid was generated with the amino-terminal histidine-tagged BBK07 driven by the polyhedrin promoter. Transfection and protein expression were performed in Sf9 cells according to the manufacturer's instructions. The protein was purified using nickel resins (Probond system; Invitrogen) to 95% purity, as assessed by SDS-PAGE analysis. The truncated BBK07 encompassing the N-terminal half of the protein (BBK07N) was expressed in and purified from Escherichia coli with a six-histidine tag as described previously (9). B. burgdorferi FlaB was purified as a recombinant protein, as described earlier (10).
Peptide synthesis.
A total of 23 overlapping peptides (P1 to P23) encompassing the mature BBK07 protein were commercially synthesized (Sigma). The peptides were 14 amino acids in length, with an overlap of 4 amino acids on the amino- and carboxyl-terminal sides. The C6 peptide (CMKKDDQIAAAMVLRGMAKDGQFALK) (22) and the pepC10 peptide (PVVAESPKKP) (25) were also commercially synthesized (GenScript). All peptides were labeled with biotin at the amino terminus and were dissolved in dimethyl sulfoxide, aliquoted, and stored at −20°C until use.
Serum.
Human serum samples used in the line blot assay were provided by Genzyme Virotech (Germany). The samples included sera collected from a total of 100 LD patients and 70 control individuals, including healthy donors and those with conditions of pregnancy, primary Epstein Barr virus infection, related spirochete infection (syphilis), and inflammatory complications (autoimmune diseases or rheumatoid factor). For ELISA studies, the healthy control sera included 25 samples from healthy human individuals who tested negative for LD, which were provided by the Centers for Disease Control and Prevention (Atlanta, GA) and by Marylou Breitentein (Yale University). These control sera were used to define the cutoff value in each assay (mean plus three standard deviations [SD]). Additional infected-serum panels were provided by the CDC. Serum panel 1 consisted of 33 serum samples collected from human patients diagnosed with LD between 2 weeks and 13 years after the onset of symptoms. Serum panel 2 consisted of 60 samples collected from 20 culture-positive patients who presented to medical doctors with erythema migrans. Serum was collected at the first doctor visit after the onset of disease (day 0) and at subsequent visits 10 and 20 days thereafter. Groups of five mice were infected with B. burgdorferi clinical isolates B408, B491, B500, B515, BL203, and BL206 (104 cells/mouse), as detailed previously (37). Two weeks after inoculation, serum samples from individual animals were collected and stored separately. Serum from a representative mouse infected with a specific isolate was used for the immunoblot assay. Canine serum samples were provided by the Synbiotics Corporation and consisted of control and infected samples, as defined by immunoblotting analysis using B. burgdorferi lysates.
ELISA and immunoblotting.
For ELISA, antigens were diluted in 50 mM carbonate-bicarbonate buffer, pH 9.6, and applied to MaxiSorp microtiter plates (Nunc). Recombinant BBK07, BBK07N, and B. burgdorferi lysate were applied at 100 ng per well and synthetic peptides (P1 to P23) (see Table 2) at 500 ng per well. BBK07 peptides and the C6 peptide were applied directly on ELISA plates at 500 ng per well. Although the BBK07 and C6 peptides were biotinylated at the N terminus, the use of streptavidin-coated plates did not influence the sensitivity of the assay (data not shown). However, in experiments using the pepC10 peptide, the plates were first coated with 400 ng per well of streptavidin (MP Biomedicals) before the addition of the peptide, in order to enhance sensitivity. The mean optical density (OD) value for the control sera plus three SD was considered the cutoff value to determine positivity in each sample. The plates were blocked using 1% bovine serum albumin (BSA) in phosphate-buffered saline with 0.05% Tween 20 (PBS-T), and plates were washed extensively with PBS-T between all steps. Human and canine sera were diluted in 1:200 or 1:10, respectively, in 1% BSA in PBS-T. Immunoblotting were performed using nitrocellulose membranes blocked with 5% skim milk in PBS-T, using murine serum dilutions from 1:200 to 1:2,000. Secondary antibodies against IgG conjugated to horseradish peroxidase were used with the following dilutions: goat anti-mouse IgG, 1:10,000; goat anti-human IgG, 1:5,000; goat anti-human IgM, 1:20,000; and goat anti-dog antibody, 1:2,000 (KPL). All steps were carried out either for 1 h at 25°C or overnight at 4°C. ELISA results were quantified using SureBlue tetramethylbenzidine (TMB) substrate (KPL). Before the optical density was read in a microplate spectrophotometer, the ELISA reactions were terminated with TMB stop solution (KPL), except for in the canine ELISA experiments, where the reactions were stopped using a 1% sodium dodecyl sulfate solution.
For the line blot assay, recombinant protein was transferred to nitrocellulose membranes by a microdispensing method as described previously (14). Strips were incubated with human sera (dilution, 1:100), and the binding of specific antibodies was detected by using alkaline phosphatase-conjugated goat anti-human IgG serum (1:100; Genzyme Virotech, Germany). Immunoreactive bands were visualized with the addition of tetramethylbenzidine substrate. Cutoffs were based on a standardized band intensity scale (version VI 0803093; Genzyme Virotech GmbH) of 0 to 5, with 5 as the most intense and 0 as no band visible. Line blots were considered positive if the band intensity was >2.
Statistical analyses.
Statistical analyses were performed using MedCalc (MedCalc Software, Mariakerke, Belgium). The diagnostic performance of each antigen was compared pairwise using the area under the curve (AUC) from receiver operator characteristic (ROC) analysis by the method of DeLong et al. (11). A serum sample was considered positive in parallel testing if it reacted positively in either or both tests.
RESULTS
BBK07 is highly sensitive and specific against B. burgdorferi infection in North American and European patients.
We have recently shown that an amino-terminal fragment of BBK07 (BBK07N) could be used for detection of human LD in North America (9). To test if BBK07 immunoreactivity was conserved in European patients, recombinant BBK07N was tested using a line blot assay which is widely used for serodiagnosis of LB in Europe (17). To assess the specificity of BBK07-based diagnosis, a set of German and U.S. patients with diagnosed LD infection, as well as sera from healthy individuals, pregnant women, and patients with other conditions, including autoimmune diseases and syphilis, were compared in a line blot assay, and the results were scored using a standardized band intensity scale. While BBK07 was highly sensitive (90%) in detecting B. burgdorferi infection in serum from the United States, the German patients with LB or other diseases or healthy controls showed minor reactivity (Table 1). The observed lower sensitivity of BBK07 reactivity in European LB patient sera likely results from the infection with other B. burgdorferi sensu lato species. Overall, these data underscore the potential sensitivity and specificity of BBK07-based diagnosis of B. burgdorferi infection in U.S. patients.
TABLE 1.
Assessment of serodiagnotic potential of BBK07N in a line blot assay
| Serum panela | No. positive/total (% positive) in BBK07N line blot assay |
|---|---|
| U.S. sera, Lyme disease IgG positive | 18/20 (90) |
| German sera | |
| Early Lyme disease | 1/20 (5) |
| Neuroborreliosis | 1/20 (5) |
| Lyme arthritis | 2/20 (10) |
| ACA | 4/20 (20) |
| Control sera | |
| Healthy blood donors | 0/20 (0) |
| EBV | 0/10 (0) |
| Syphilis infections | 0/10 (0) |
| Pregnant women | 0/10 (0) |
| Autoimmune diseases | 1/10 (10) |
| Rheumatoid factor | 0/10 (0) |
Human serum samples were analyzed by a line blot assay with various amounts of the recombinant fragment BBK07N. ACA, acrodermatitis chronica atrophicans; EBV, primary Epstein-Barr virus infection.
A BBK07-specific immune response is elicited during murine infection with several clinical isolates of B. burgdorferi sensu stricto.
In order to further characterize the diagnostic potential of BBK07, we measured its conservation and immunogenicity in multiple clinical strains which were isolated from human patients from North America (37). These isolates have been previously characterized by restriction fragment length polymorphism analysis. Isolates B491, B515, BL203, and BL206 have a genotype associated with invasive disease, while B408 and B500 are likely to cause less invasive infections (37). The B. burgdorferi clinical isolates were grown in BSK medium and injected into mice, and serum was collected at 2 or 3 weeks following infection. Infected murine sera were used to probe equal amounts of recombinant BBK07N or FlaB protein by immunoblot analysis (Fig. 1). A BBK07-specific antibody response was detectable in mice infected with the B. burgdorferi B31 laboratory isolate A3, as well as all clinical isolates. Although we lack human data, these murine studies raise the possibility that BBK07-based serodiagnosis may be able to detect LD infection with multiple clinical isolates of B. burgdorferi sensu stricto.
FIG. 1.
Conservation of BBK07 immunoreactivity in clinical B. burgdorferi isolates. Mice were infected with several B. burgdorferi isolates from infected human patients in North America. Sera from the infected mice were collected after 2 weeks of infection and used to probe 200 ng of recombinant BBK07N or FlaB proteins to test for an immune response against BBK07. A BBK07-specific immune response was detected in the widely used laboratory strain A3, as well as in each mouse infected with a human isolate (297 through BL206). No antibodies were detected using normal mouse serum (NMS).
Identification of immunogenic epitopes of BBK07.
We next mapped the immunodominant epitopes of BBK07 using human sera. To accomplish this, a library consisting of 23 overlapping peptides was synthesized, covering the full-length mature BBK07 (Table 2). Each peptide was tested for ELISA reactivity in serum samples from 25 healthy donors and 33 patients diagnosed with LD from serum panel 1. Given the number of peptides and the limited supply of serum, all peptides were screened once with each serum sample, and the most immunogenic peptides were repeated in three independent experiments to validate the results. The cutoff value was set at the mean plus three SD of results for the healthy controls for each antigen. The results indicated that amino-terminal peptides P1 (CKWHVDNPIDEATA), P5 (ITKLTPEELENLAK), and P7 (EKSKKEIEDQKNTK) harbored the most immunodominant epitopes recognized by the infected sera (Fig. 2 A). Samples were considered positive if they exceeded the cutoff value in at least two out of three independent ELISAs (Fig. 2B). We speculated that additional testing of other peptides was unlikely to significantly increase sensitivity, as the serum samples that were reactive to other peptides were also reactive to peptide P1, P5, or P7. We therefore focused our attention on these three peptides, and repeated measurements confirmed that among peptides P1, P5 and P7, P5 carried the most immunogenic epitopes and was able to distinguish 24% (8 out of 33) of the LD-diagnosed patient sera from control sera. In order to assess if the combination of the three most immunogenic peptides increased the sensitivity of the assay, the peptides were mixed and tested against the serum panel (Fig. 2B). The results indicated that the combination of three peptides had a synergistic effect and increased the diagnostic accuracy to 52% (P1P5P7; 17 out of 33), significantly increasing the test performance by receiver operator characteristic (ROC) analysis (P < 0.05).
TABLE 2.
BBK07 peptide library amino acid sequences
| Peptide name | Amino acid sequence |
|---|---|
| P1 | CKWHVDNPIDEATA |
| P2 | EATAESKSALTSVD |
| P3 | TSVDQVLDEISEAT |
| P4 | SEATGLSSEKITKL |
| P5 | ITKLTPEELENLAK |
| P6 | NLAKEAQDDSEKSK |
| P7 | EKSKKEIEDQKNTK |
| P8 | KNTKESKNIEVKDT |
| P9 | VKDTPRLIKLIKNS |
| P10 | IKNSSEKIDSVFQT |
| P11 | VFQTLINIGYNATY |
| P12 | NATYAAKSNLKNGL |
| P13 | KNGLKMVKLLDELL |
| P14 | DELLKISVSSNGDK |
| P15 | NGDKSTQKYNELKT |
| P16 | ELKTVVNRFNAENS |
| P17 | AENSAIKVPLENGS |
| P18 | ENGSKIEAKKCIKT |
| P19 | CIKTLMTNVETYFK |
| P20 | TYFKGVSTELKDKK |
| P21 | KDKKDDKYTKILAA |
| P22 | ILAALSEAANKIEN |
| P23 | KIENAAMAIHLCFNN |
FIG. 2.
Identification of immunodominant epitopes of BBK07. (A) Epitope mapping of BBK07 using infected human sera. A library of 23 overlapping BBK07 peptides was screened for reactivity in sera from 33 patients diagnosed with Lyme disease (LD) or from 25 healthy donors. Microtiter wells were coated with individual BBK07 peptides (500 ng/well) and incubated with individual patient or healthy donor serum (1:200 dilution). The signal was generated using goat secondary antibodies against human IgG conjugated to horseradish peroxidase at a dilution of 1:5000. The cutoff value for each peptide was set as the mean value for the healthy control serum panel plus three SD. Bars represent the percentage of diagnosed serum samples above the cutoff value. (B) A peptide combination significantly enhances BBK07-based serodiagnostic ability compared to that with individual peptides. Peptides P1, P5, and P7 (hatched bars) and all three peptides combined (P1P5P7; solid bars) were tested in three independent ELISA measurements for reactivity with human serum. Serum samples were considered positive if the result was above the cutoff value in at least two out of three experiments.
A combination of BBK07 peptides provides greater diagnostic accuracy than recombinant BBK07.
We have demonstrated that a combination of BBK07 peptides (Fig. 2) and a recombinant amino-terminal fragment of BBK07 (9) are effective in LD diagnosis. To assess whether the presence of conformational epitopes could improve the sensitivity of BBK07-based serodiagnosis, we produced recombinant full-length BBK07 in insect cells using a baculovirus expression system. The full-length protein was tested against serum panel 1 as detailed above, and its reactivity was compared to those of the recombinant amino-terminal fragment BBK07N and the BBK07 peptide combination. B. burgdorferi lysate was also included as a positive control. Three independent measurements were performed, which produced similar results (data not shown), and results of a representative experiment are shown (Fig. 3). While we expected that the full-length BBK07 (which distinguished 13 out of 33 patient sera from control sera [39%]) should provide more epitopes for antibody recognition, the amino-terminal fragment BBK07N (14 out of 33 [42%]) showed no significant reduction in diagnostic accuracy, as assessed by statistical analysis (P = 0.181). However, the peptide combination (17 out of 33 [52%]) displayed a diagnostic accuracy superior to that of either BBK07N or full-length BBK07. Statistical analyses further indicated that there was a significant improvement in test performance for the peptide combination compared to the full-length protein (P = 0.048) but not for the peptide combination compared to the recombinant fragment BBK07N (P = 0.252).
FIG. 3.
Immunoreactivities of full-length or truncated BBK07 proteins and peptides to individual Lyme disease patient sera. Microtiter wells were coated with 100 ng of B. burgdorferi lysate (Bb lysate), recombinant full-length BBK07 protein (BBK07), the N-terminal BBK07 fragment (BBK07N), or a combination of immunodominant peptides P1, P5, and P7 (P1P5P7) (500 ng each). The immunoreactivities of individual serum samples from healthy donors (black circles) and from patients diagnosed with LD (white circles) were measured by ELISA. The horizontal dashed lines indicates the cutoff values calculated for the antigens from the values for sera from healthy donors. Values higher than 3 are represented on the graph as 3 for clarity of presentation.
BBK07 peptide-based diagnosis is effective during early infection.
We next assessed if BBK07-based diagnosis is effective in early LD infection. Since the peptide combination displayed the highest BBK07-based serodiagnostic sensitivity, this was used in the study. Serum samples (serum panel 2) were collected from patients displaying erythema migrans on their first visit to the physician. Biopsy specimens from all patients were later culture positive for B. burgdorferi, and serum from each patient was collected again at 10 and 20 days after the first visit (day 0) to the doctor. Note that the serum is first collected from the patients upon their first visit to a physician and not at the time of initial exposure. Therefore, while it is not known how long after infection the serum samples were collected, the appearance of erythema migrans on the first visit to a physician makes it likely to be within the first few weeks of B. burgdorferi infection (36). The control sera from serum panel 1 were used to define the cutoff for each antigen as detailed above, and each antigen was used to test serum panel 2 in three independent measurements (Tables 3 and 4). The BBK07 peptide combination P1P5P7 had a similar diagnostic accuracy in detecting IgG at day 10 as observed using serum panel 1 (10 out of 20 [50%]) (Fig. 4). To further assess the effectiveness of the P1P5P7 peptide combination in the diagnosis of early disease, it was tested for both IgM and IgG reactivity using appropriate controls, such as the OspC peptide pepC10 for IgM (25) and the VlsE peptide C6 for IgG detection (22). The results show that, compared to BBK07, the peptides pepC10 and C6 showed higher overall sensitivities in detecting IgM and IgG, respectively (Tables 3 and 4). However, a subset of patients who displayed negative reactivity to the pepC10 and C6 peptides reacted positively against P1P5P7 combination (Tables 3 and 4). These data indicate heterogeneity in the immune responses of the tested patients and suggest the potential use of BBK07 peptides for enhancing the diagnostic accuracy of LD detection.
TABLE 3.
Individual patient IgM reactivity against BBK07 (P1P5P7) and OspC (pepC10) for peptide-based serodiagnosis of early LD
| Patient | IgM reactivitya against: |
|||||
|---|---|---|---|---|---|---|
| P1P5P7 |
pepC10 |
|||||
| Initial visit (0/20, 0%) | Day 10 (4/20, 20%) | Day 20 (2/20, 10%) | Initial visit (3/20, 15%) | Day 10 (10/20, 50%) | Day 20 (9/20, 45%) | |
| 1 | − | + | + | − | − | − |
| 2 | − | − | − | − | + | + |
| 3 | − | − | − | − | + | + |
| 4 | − | − | − | − | − | − |
| 5 | − | − | − | − | − | − |
| 6 | − | − | − | − | − | − |
| 7 | − | − | − | − | − | − |
| 8 | − | − | − | − | + | + |
| 9 | − | − | − | − | − | − |
| 10 | − | − | − | + | + | + |
| 11 | − | − | − | − | − | − |
| 12 | − | − | − | + | + | + |
| 13 | − | + | − | − | + | + |
| 14 | − | + | − | − | + | + |
| 15 | − | − | − | + | + | + |
| 16 | − | + | − | − | + | − |
| 17 | − | − | − | − | − | − |
| 18 | − | − | + | − | − | − |
| 19 | − | − | − | − | − | − |
| 20 | − | − | − | − | + | + |
Boldface indicates samples negative for pepC10 but positive for P1P5P7.
TABLE 4.
Individual patient IgG reactivity against BBK07 (P1P5P7) and VlsE (C6) for peptide-based serodiagnosis of early LD
| Patient | IgM reactivitya against: |
|||||
|---|---|---|---|---|---|---|
| P1P5P7 |
C6 |
|||||
| Initial visit (3/20, 15%) | Day 10 (10/20, 50%) | Day 20 (8/20, 40%) | Initial visit (11/20, 55%) | Day 10 (14/20, 70%) | Day 20 (15/20, 75%) | |
| 1 | − | + | + | + | + | + |
| 2 | − | + | − | + | + | + |
| 3 | − | − | − | − | − | − |
| 4 | − | − | − | − | − | − |
| 5 | − | − | − | − | + | + |
| 6 | − | − | − | + | + | + |
| 7 | − | − | − | + | − | − |
| 8 | − | − | − | − | + | + |
| 9 | + | + | + | − | − | + |
| 10 | − | − | − | + | + | + |
| 11 | − | + | − | + | + | + |
| 12 | − | − | − | + | + | + |
| 13 | − | + | + | − | + | + |
| 14 | − | + | + | − | − | − |
| 15 | − | + | + | + | + | + |
| 16 | − | + | + | − | + | + |
| 17 | + | + | + | + | + | + |
| 18 | − | + | + | − | − | − |
| 19 | − | − | − | + | + | + |
| 20 | + | − | − | + | + | + |
Boldface indicates samples negative for C6 but positive for P1P5P7.
FIG. 4.
BBK07 antibody development in LD patients during early infection. (A) Early immunoreactivity to BBK07 peptides. The BBK07 peptide combination (P1P5P7) was assessed for immunoreactivity against the same cohort of human patients during early LD. Sera from a group of 20 patients experiencing the early symptoms of LD were collected during the initial doctor visit (day 0) and two subsequent visits (days 10 and 20) and were used in the assay. Sera from 20 healthy controls were used to establish a cutoff value, and results of a representative experiment are shown. (B) Graphical representation of the data shown in panel A. The samples were tested in three independent ELISA measurements and were considered positive if above the cutoff value in at least two out of three measurements. Totals of 15%, 50%, and 40% of patients were positive for days 0, 10, and 20, respectively.
BBK07 peptide-based diagnosis is highly effective in canine LD.
As BBK07 immunoreactivity is an effective serodiagnostic marker for the detection of B. burgdorferi infection in mice and humans (9), we next assessed if the antigen could be used to detect LD in dogs that are routinely screened for LD infection. A panel of 55 uninfected and 46 B. burgdorferi-infected canine serum samples was assessed using the BBK07 peptide combination P1P5P7 in an ELISA, as described above. The assay was repeated with similar results, and results of a representative experiment are shown (Fig. 5). The results show that like in humans, the peptides from BBK07 are also effective (35 out of 46 [76%]) for the detection of LD in dogs.
FIG. 5.
The immunodominant BBK07 peptide as a potential antigenic target for serodiagnosis of canine Lyme disease. Sera from healthy dogs (black circles; n = 55) and ones diagnosed with Lyme disease (white circles; n = 46) were used to assess immunoreactivity of the BBK07 peptide combination P1P5P7 by ELISA. Optical density values for control dog serum were used as controls to set the cutoff value for the antigen (horizontal dashed line).
DISCUSSION
The identification of new immunodominant antigens could improve the current laboratory diagnosis of LD. BBK07, a surface-exposed B. burgdorferi protein that is induced in vivo during mammalian infection, has been indicated as a potential serodiagnostic marker for LD (9). To further evaluate the potential of BBK07 as a serodiagnostic marker, we assessed its sensitivity and specificity using diverse sets of serum collected from the United States and Germany, where samples were chosen from patients with diagnosed LD and several relevant and cross-reactive diseases and from healthy controls. When mice were infected with seven different clinical isolates of B. burgdorferi originally cultured from human patients, each prompted a detectable antibody response to BBK07 suggesting wide conservation of BBK07 immunogenicity among clinical isolates. Using a library of 23 overlapping peptides and a human LD serum panel, we further identified the most immunodominant epitopes of BBK07 and showed that the combination of three amino-terminal peptides is a significantly more effective serodiagnostic marker than the full-length antigen. We also showed that this peptide combination can be used as a potential component of a serodiagnostic marker in early stages of the disease, effectively diagnosing some patients who lack reactivity to OspC- or VlsE-derived peptides, such as pepC10 or C6, respectively. Finally, we showed that these results may be applicable to other, nonhuman species at high risk of B. burgdorferi infection, as peptides from BBK07 were able to effectively detect LD in canine serum samples. As immunoreactivity could potentially vary across mammalian species, the effectiveness of a BBK07 peptide to diagnose canine LD in our preliminary study also testifies to the future use of BBK07-based diagnosis in companion animals. It has been suggested that dogs may serve as sentinels for LD, as they can be at a higher risk of tick exposure (23). Taken together, our data identified the most immunodominant epitopes of BBK07 and highlighted the usefulness of a BBK07 peptide combination as a potential component of a highly sensitive and specific test for human and canine LD.
The effectiveness of some B. burgdorferi antigens, including VlsE (18), OspC (12), and BmpA (33), has been reduced by sequence variation in the bacterial population. Our data indicate that BBK07 immunoreactivity is detectable across the B. burgdorferi isolates present in North America but not in European LD patients. Although the BBK07 gene is highly conserved in B. burgdorferi sensu stricto isolates in United States, the linear plasmid carrying the BBK07 gene or an ortholog thereof is absent in major B. burgdorferi sensu lato strains prevalent in Europe. Therefore, while BBK07 diagnosis is unlikely to be effective in Europe, reactivity to BBK07 or similar antigens absent in other B. burgdorferi sensu lato species could aid physicians or researchers in differentiating between individuals infected with B. burgdorferi or the European strains (6, 38).
Our previous work indicated that the amino-terminal half of BBK07 was immunogenic in humans and accessible to protease degradation on the surface of B. burgdorferi cells (9). This region is likely to be the major target of the host immune response, as our primary screen of the peptide library also showed that the most immunodominant peptides of BBK07 were located on the amino terminus of the protein. Two of the most effective BBK07 peptide antigens, P5 (amino acids 56 through 69) and P7 (amino acids 76 through 89), are situated only six amino acids apart on the amino-terminal region of BBK07. Their close proximity suggests that a longer peptide, encompassing P5, P6, and P7, could effectively harness the linear epitopes of each on a single molecule. In addition, the longer peptide could form limited conformational epitopes present in this region of the protein, potentially increasing sensitivity further. Moreover, BBK07 peptide diagnosis could further be enhanced by improvements in antibody detection techniques. In our study, we utilized a simple ELISA, absorbing the peptides directly on the plastic surface in a high-pH buffer. The sensitivity of the test is then dependent on the ability of the plate to bind and present the peptide. In our assay, each peptide was biotinylated on the amino terminus, but streptavidin-coated plates did not increase the sensitivity of the test. Chemical conjugation to solid surfaces or fusions with carrier proteins have been used in the past to improve peptide binding and display and could be applied to further enhance sensitivity of BBK07 peptide-based diagnosis (15). A recent study demonstrated enhanced sensitivity of LD detection using a luciferase immunoprecipitation system (8). This technique is performed entirely in the liquid phase and has a broad dynamic detection range. The highest sensitivity was achieved using a synthetic peptide hybrid containing epitopes from the B. burgdorferi proteins OspC and VlsE. Our data show that BBK07 peptides were able to detect an antibody response in several patients whose sera did not react with the VlsE peptide C6 or the OspC peptide. The mechanism by which C6- or OspC-negative individuals might respond to BBK07 peptides currently remains unknown. However, this could potentially result from the differential expression of spirochete surface antigens in certain humans or variation of individual immune responses against particular peptides, among other possibilities.
In conclusion, our data show that select peptides from BBK07 are more effective antigens than the full-length BBK07 protein, have the ability to detect both early and late human infections, and can detect a variety of B. burgdorferi infectious isolates. These factors, combined with their unique ability to detect B. burgdorferi in patients not reactive to C6 or pepC10 peptides, suggest that epitopes from BBK07 could be used as components of a more efficient sensitive and specific serodiagnostic test for LD.
Acknowledgments
This work was supported by Public Health Service grants AR055323 and AI080615 (to U.P.) and AR041511 (to I.S.) from the National Institutes of Health.
We are grateful to Martin Schriefer for access to the CDC Lyme serum panel, collected from patients with diagnosed Lyme disease as well as from healthy individuals from areas where the disease is not endemic.
Footnotes
Published ahead of print on 22 December 2010.
REFERENCES
- 1.Aguero-Rosenfeld, M. E., J. Nowakowski, D. F. McKenna, C. A. Carbonaro, and G. P. Wormser. 1993. Serodiagnosis in early Lyme disease. J. Clin. Microbiol. 31:3090-3095. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Aguero-Rosenfeld, M. E., G. Wang, I. Schwartz, and G. P. Wormser. 2005. Diagnosis of Lyme borreliosis. Clin. Microbiol. Rev. 18:484-509. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Anonymous. 1995. Recommendations for test performance and interpretation from the Second National Conference on Serologic Diagnosis of Lyme Disease. Morb. Mortal. Wkly. Rep. 44:590-591. [PubMed] [Google Scholar]
- 4.Bacon, R. M., et al. 2003. Serodiagnosis of Lyme disease by kinetic enzyme-linked immunosorbent assay using recombinant VlsE1 or peptide antigens of Borrelia burgdorferi compared with 2-tiered testing using whole-cell lysates. J. Infect. Dis. 187:1187-1199. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Bacon, R. M., K. J. Kugeler, and P. S. Mead. 2008. Surveillance for Lyme disease—United States, 1992-2006. MMWR Surveill. Summ. 57:1-9. [PubMed] [Google Scholar]
- 6.Barbour, A. G., et al. 2008. A genome-wide proteome array reveals a limited set of immunogens in natural infections of humans and white-footed mice with Borrelia burgdorferi. Infect. Immun. 76:3374-3389. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Brissette, C. A., et al. 2010. The borrelial fibronectin-binding protein RevA is an early antigen of human Lyme disease. Clin. Vaccine Immunol. 17:274-280. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Burbelo, P. D., et al. 2010. Rapid, simple, quantitative, and highly sensitive antibody detection for Lyme. Dis. Clin. Vaccine Immunol. 17:904-909. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Coleman, A. S., and U. Pal. 2009. BBK07, a dominant in vivo antigen of Borrelia burgdorferi, is a potential marker for serodiagnosis of Lyme disease. Clin. Vaccine Immunol. 16:1569-1575. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Cox, D. L., et al. 1996. Limited surface exposure of Borrelia burgdorferi outer surface lipoproteins. Proc. Natl. Acad. Sci. U. S. A. 93:7973-7978. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.DeLong, E. R., D. M. DeLong, and D. L. Clarke-Pearson. 1988. Comparing the areas under two or more correlated receiver operating characteristic curves: a nonparametric approach. Biometrics 44:837-845. [PubMed] [Google Scholar]
- 12.Earnhart, C. G., E. L. Buckles, J. S. Dumler, and R. T. Marconi. 2005. Demonstration of OspC type diversity in invasive human Lyme disease isolates and identification of previously uncharacterized epitopes that define the specificity of the OspC murine antibody response. Infect. Immun. 73:7869-7877. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Elias, A. F., et al. 2002. Clonal polymorphism of Borrelia burgdorferi strain B31 MI: implications for mutagenesis in an infectious strain background. Infect. Immun. 70:2139-2150. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Goettner, G., et al. 2005. Improvement of Lyme borreliosis serodiagnosis by a newly developed recombinant immunoglobulin G (IgG) and IgM line immunoblot assay and addition of VlsE and DbpA homologues. J. Clin. Microbiol. 43:3602-3609. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Gomara, M. J., and I. Haro. 2007. Synthetic peptides for the immunodiagnosis of human diseases. Curr. Med. Chem. 14:531-546. [DOI] [PubMed] [Google Scholar]
- 16.Heikkila, T., et al. 2002. Recombinant BBK32 protein in serodiagnosis of early and late Lyme borreliosis. J. Clin. Microbiol. 40:1174-1180. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Hunfeld, K. P., and P. Kraiczy. 2009. When is the best time to order a Western blot and how should it be interpreted? Curr. Probl. Dermatol. 37:167-177. [DOI] [PubMed] [Google Scholar]
- 18.Iyer, R., J. M. Hardham, G. P. Wormser, I. Schwartz, and S. J. Norris. 2000. Conservation and heterogeneity of vlsE among human and tick isolates of Borrelia burgdorferi. Infect. Immun. 68:1714-1718. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Kaiser, R., and S. Rauer. 1999. Advantage of recombinant borrelial proteins for serodiagnosis of neuroborreliosis. J. Med. Microbiol. 48:5-10. [DOI] [PubMed] [Google Scholar]
- 20.Lawrenz, M. B., et al. 1999. Human antibody responses to VlsE antigenic variation protein of Borrelia burgdorferi. J. Clin. Microbiol. 37:3997-4004. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Ledue, T. B., M. F. Collins, and W. Y. Craig. 1996. New laboratory guidelines for serologic diagnosis of Lyme disease: evaluation of the two-test protocol. J. Clin. Microbiol. 34:2343-2350. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Liang, F. T., et al. 1999. Sensitive and specific serodiagnosis of Lyme disease by enzyme-linked immunosorbent assay with a peptide based on an immunodominant conserved region of Borrelia burgdorferi vlsE. J. Clin. Microbiol. 37:3990-3996. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Lindenmayer, J. M., D. Marshall, and A. B. Onderdonk. 1991. Dogs as sentinels for Lyme disease in Massachusetts. Am. J. Public Health 81:1448-1455. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Magnarelli, L. A., E. Fikrig, S. J. Padula, J. F. Anderson, and R. A. Flavell. 1996. Use of recombinant antigens of Borrelia burgdorferi in serologic tests for diagnosis of Lyme borreliosis. J. Clin. Microbiol. 34:237-240. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Mathiesen, M. J., et al. 1998. Peptide-based OspC enzyme-linked immunosorbent assay for serodiagnosis of Lyme borreliosis. J. Clin. Microbiol. 36:3474-3479. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Mueller, M., et al. 2006. Identification of Borrelia burgdorferi ribosomal protein L25 by the phage surface display method and evaluation of the protein's value for serodiagnosis. J. Clin. Microbiol. 44:3778-3780. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Padula, S. J., F. Dias, A. Sampieri, R. B. Craven, and R. W. Ryan. 1994. Use of recombinant OspC from Borrelia burgdorferi for serodiagnosis of early Lyme disease. J. Clin. Microbiol. 32:1733-1738. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Pal, U., et al. 2004. OspC facilitates Borrelia burgdorferi invasion of Ixodes scapularis salivary glands. J. Clin. Invest. 113:220-230. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Piesman, J., and L. Eisen. 2008. Prevention of tick-borne diseases. Annu. Rev. Entomol. 53:323-343. [DOI] [PubMed] [Google Scholar]
- 30.Ramamoorthy, R., and M. T. Philipp. 1998. Differential expression of Borrelia burgdorferi proteins during growth in vitro. Infect. Immun. 66:5119-5124. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Rasiah, C., S. Rauer, G. S. Gassmann, and A. Vogt. 1994. Use of a hybrid protein consisting of the variable region of the Borrelia burgdorferi flagellin and part of the 83-kDa protein as antigen for serodiagnosis of Lyme disease. J. Clin. Microbiol. 32:1011-1017. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Rauer, S., N. Spohn, C. Rasiah, U. Neubert, and A. Vogt. 1998. Enzyme-linked immunosorbent assay using recombinant OspC and the internal 14-kDa flagellin fragment for serodiagnosis of early Lyme disease. J. Clin. Microbiol. 36:857-861. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.Roessler, D., U. Hauser, and B. Wilske. 1997. Heterogeneity of BmpA (P39) among European isolates of Borrelia burgdorferi sensu lato and influence of interspecies variability on serodiagnosis. J. Clin. Microbiol. 35:2752-2758. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Schwan, T. G., W. Burgdorfer, and C. F. Garon. 1988. Changes in infectivity and plasmid profile of the Lyme disease spirochete, Borrelia burgdorferi, as a result of in vitro cultivation. Infect. Immun. 56:1831-1836. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35.Simpson, W. J., M. E. Schrumpf, and T. G. Schwan. 1990. Reactivity of human Lyme borreliosis sera with a 39-kilodalton antigen specific to Borrelia burgdorferi. J. Clin. Microbiol. 28:1329-1337. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36.Steere, A. C. 2006. Lyme borreliosis in 2005, 30 years after initial observations in Lyme Connecticut. Wien. Klin. Wochenschr. 118:625-633. [DOI] [PubMed] [Google Scholar]
- 37.Wang, G., et al. 2002. Disease severity in a murine model of Lyme borreliosis is associated with the genotype of the infecting Borrelia burgdorferi sensu stricto strain. J. Infect. Dis. 186:782-791. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38.Wilske, B., V. Fingerle, and U. Schulte-Spechtel. 2007. Microbiological and serological diagnosis of Lyme borreliosis. FEMS Immunol. Med. Microbiol. 49:13-21. [DOI] [PubMed] [Google Scholar]